基于麻雀搜索算法的降水量预测  

Precipitation prediction based on sparrow search algorithm

作　　者：李淼 宇世航[1] LI Miao;YU Shihang(School of Science,Qiqihar University,Qiqihar 161006,China)

机构地区：[1]齐齐哈尔大学理学院,黑龙江齐齐哈尔161006

出　　处：《高师理科学刊》2024年第5期28-34,共7页Journal of Science of Teachers＇College and University

基　　金：黑龙江省省属高等学校基本科研业务费科研项目(145209138);齐齐哈尔大学研究生创新科研项目(QUZLTS_CX2023051);齐齐哈尔大学教育科学研究项目(GJQTYB202105)。

摘　　要：为提高对非线性和时序性降水量数据的准确预报,建立了基于麻雀搜索算法的组合预测模型.利用经验模态分解挖掘数据多维度特征,使用长短期记忆人工网络对数据进行预测,结合麻雀搜索算法对预测模型的超参数进行优化,提高了经验模态分解-长短期记忆人工网络模型网格化寻参的效率和预测精度.实证结果表明,与单一的长短期记忆人工网络模型和经验模态分解-长短期记忆人工网络模型相比,经过优化后的基于麻雀搜索算法的组合预测模型的性能和预测效果更好,其各类误差均有所降低,具有实际意义.To improve the accurate prediction of nonlinear and temporal precipitation data,a combined prediction model based on the sparrow search algorithm was established.Empirical mode decomposition was used to mine multidimensional features of the data,and long short-term memory artificial networks were used to predict the data,the hyper parameters of the prediction model were optimized using the sparrow search algorithm,the efficiency and prediction accuracy of grid based parameter search in empirical mode decomposition-long short term memory artificial network models is improved.Empirical results show that compared with a single long short term memory artificial network model and the empirical mode decomposition-long short term memory artificial network model,the optimized combination prediction model based on the sparrow search algorithm has better performance and prediction effect,and all types of errors are reduced,which has practical significance.

关 键 词：降水量预测 麻雀搜索算法 经验模态分解-长短期记忆人工网络模型 

分 类 号：O29[理学—应用数学]